Anxiety disorders are the most common psychiatric disorders among youth, with lifetime prevalence ranging between 18-20% of the general population. Among anxiety disorders, social anxiety disorder (SAD) affects 8% of all youth, resulting in significant short and long-term impairment, including increased likelihood of substance abuse, limited academic achievement, attenuated occupational functioning, and impaired or missing social relationships. Emerging data suggest that interventions that include social skills training, formal peer generalization sessions, and homework assignments as part of an overall treatment strategy show enhanced efficacy when compared to interventions without these components - the latter two elements which are designed to enhance skill generalization. Two critical treatment elements (peer generalization, homework assignments) are difficult to implement in traditional clinical settings, limiting optimal dissemination to youth in need of thes services in different settings (e.g., at school, outpatient, or community facilities). In our recenly completed Phase I STTR, we developed and validated an interactive virtual environment (VE) to solve the need for intensive behavioral practice opportunities that are critical for skill generalization. Our VE, known as Pegasys-VR, allowed intensive practice of social skills without the need for formal peer group activities (in-clinic solution) or intensive parental involvement (at-home solution). Our results indicated that implementing a VE environment into the SET-C program was accessible, credible and feasible for the parents, clinicians and children who participated in the trial. Furthermore, examination of within group changes indicated statistically significant improvement in SAD symptoms. Given the success of the Phase I, and the need to increase the automation of the system to increase its acceptance and adoption among clinicians and socially anxiety youth, our Phase II STTR study will have the following goals: incorporate an artificial intelligence natural language system, develop additional VEs for in-clinic practice, expand the homework solution to provide additional opportunities to reinforce skills acquisition and generalization practice in in-vivo settings, using serious game theory and technology, and conduct a randomized controlled trial with youth ages 7 to 12 to test whether Pegasys-VR is as effective as traditional behavioral interventions for youth with social anxiety disorder and if the effects are maintained at 3 month follow-up. If Pegasys-VR is clinically efficacious, it would offer a sustainable, cost-effective intervention that can be easily and rapidy disseminated.